Shubham Deshmukh

Shubham Deshmukh, Favin Fernandes, Amey Chavan

With the advancements in Computer vision techniques the need to classify images based on its features have become a huge task and necessity. In this project we proposed 2 models i.e. feature extraction and classification using ORB and SVM and the second is using CNN architecture. The end result of the project is to understand the concept behind feature extraction and image classification. The trained CNN model will also be used to convert it to tflite format for Android Development.

Shubham Deshmukh, Favin Fernandes, Amey Chavan et al.

The proposed shopping assistant model SANIP is going to help blind persons to detect hand held objects and also to get a video feedback of the information retrieved from the detected and recognized objects. The proposed model consists of three python models i.e. Custom Object Detection, Text Detection and Barcode detection. For object detection of the hand held object, we have created our own custom dataset that comprises daily goods such as Parle-G, Tide, and Lays. Other than that we have also collected images of Cart and Exit signs as it is essential for any person to use a cart and also notice the exit sign in case of emergency. For the other 2 models proposed the text and barcode information retrieved is converted from text to speech and relayed to the Blind person. The model was used to detect objects that were trained on and was successful in detecting and recognizing the desired output with a good accuracy and precision.

Shubham Deshmukh, Favin Fernandes, Monali Ahire et al.

In this project we propose a CNN architecture to detect anomaly and suspicious activities; the activities chosen for the project are running, jumping and kicking in public places and carrying gun, bat and knife in public places. With the trained model we compare it with the pre-existing models like Yolo, vgg16, vgg19. The trained Model is then implemented for real time detection and also used the. tflite format of the trained .h5 model to build an android classification.

Vedant Ghodke, Shubham Deshmukh, Atharva Deshpande et al.

With the growth of embedded systems, VLSI design phases complexity and cost factors across the globe and has become outsourced. Modern computing ICs are now using system-on-chip for better on-chip processing and communication. In the era of Internet-of-Things (IoT), security has become one of the most crucial parts of a System-on-Chip (SoC). Malicious activities generate abnormal traffic patterns which affect the operation of the system and its performance which cannot be afforded in a computation hungry world. SoCs have a chance of functionality failure, leakage of information, even a denial of services (DoS), Hardware Trojan Horses and many more factors which are categorized as security threats. In this paper, we aim to compare and describe different types of malicious security threats and how neural networks can be used to prevent those attacks. Spiking Neural Networks (SNN), Runtime Neural Architecture (RTNA) are some of the neural networks which prevent SoCs from attacks. Finally, the development trends in SoC security are also highlighted.